Indexical Speech across Samoan CommunitiesAuthor(s): Alessandro DurantiReviewed work(s):Source: American Anthropologist, New Series, Vol. 99, No. 2 (Jun., 1997), pp. 342-354Published by: Blackwell Publishing on behalf of the American Anthropological AssociationStable URL: http://www.jstor.org/stable/682215 .Accessed: 25/07/2012 23:35

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

.

Blackwell Publishing and American Anthropological Association are collaborating with JSTOR to digitize,preserve and extend access to American Anthropologist.

http://www.jstor.org

IN TRANSNATIONAL COMMUNITIES, speaking about space can be a way of bridging physically distant but emotionally and ethically close worlds. A comparison of the use and context of one particular Samoan expres- sion, nofo i lalo! (sit down!"), as used in a Western Sa- moan village and a suburban neighborhood in southern CalifoInia, suggests that the indexical grounding of this expression within a horizon of speciElc body orienta- tions and material artifacts not only opens it to a variety of interpretations but also makes it uready-to-use" in es- tablishing connections with a distant world that might be 1lnknown to the recipient of the message. In the IJnited States, a Samoan mother might issue an instruc- tion in Samoan regarding proper spatial orientation to children who have never been to a Samoan island and have not expenenced the world of material possessions and ethical stances that is presupposed by such an in- struction. In this context1 such a command is more than an instruction uttered by a frustrated mother who is try- ing to control restless children jumping around in the living room. It is also an attempt to evoke an interpre- tive horizon where vertical and horizontal positioning of one's body in the living space has socioethical impli- cations that, once recognized, can establish a cultural continuity that is otherwise deEled by the built environ- ment in vvhich the interaction takes place. aTalking spaceX in this way then becomes another contested ground where the battle between continuity and change can be fought. It is another way of drawing the bounda- ries of the community, this time not on paper but in the interactional fabric of live cliscourse.

Before engaging in the analysis of living space in two distant Samoan communities, I must Elrst deal with a problem in the language of anthropologists. Although

an expression used by earlier researchers as a general label for a range of local notions and practices might help us make sense of what we are experiencing, it might also stop us from seeing other, equally important practices.

Old and New Anthropological Key Expressions

Each ethnographer has a stock of anthropological key expressions, representing misleadingly simple con- cepts with the unique power to evoke endless streams of associated meanings and provide countless occa- sions for anecdotes, descriptionsX and hypotheses, reaching all the way into local or even universal theo- ries of the human mind, the social nature of things, the thingness of human action. Fwor many cultural anthro- pologists key expressions are either a single word or a pair of words, usually an oppositional set. A classic ex- ample is Clifford Geertz's (1983) use of the Arabic term nisba, which he elucidates with a long list of associated meanings and etymologically related terms. According to Geertz, to understand ntsba means to get a hold of the main instrument through which Moroccans sort people out from one another and form an idea of what it is to be a person" (1983:65). Wisba, then, becomes a key expression for understanding the Moroccan sense of person, a central focus of Geertz's brand of interpretive anthropology.

Another example is the pair ligetAbeya, roughly zknowledge/anger, passion," in Michelle Rosaldo's (1980) ethnography of the Ilongots. For Rosaldo, under- standing how the semantic domains associated with these two terms are organized, and how they are made sense of in everyday practices, means coming to terms with the Ilongot worldview and way of life, including head-hunting practices. As Daniel Rosenberg has rightly pointed out, in this kind of lexically organized representation of culture, we risk essentializing a vast

ALESSANDRO DURANTI is a professor in the Department of Anthropology, University of California, Los Angeles, CA 90095.

Amencan Anthropologist99(2):342-354. Copyright O 1997, American Anthropological Association.

ALESSANDRO DURANTI / UNIVERSITY OF CALIFORNIA LOS ANGELES

Indexisal Speeeh aeross Samoan

Communities

INDEXICAL SPEECH ACROSS SAMOAN COMMUNITIES / ALESSANDRO DURANTI 343

range of complex cultural practices in terms of one or two words (Rosenberg 1990:166). We also use language in a very limited sense, distilling a few drops of lexical material out of the multifarious and rich usages that characterize any real-life situation.

Yet key expressions have been shown to be power- ful tools for the research@r/observer. These words and (more rarely) phrases, smuggled through the invisible and yet quite real customs of anthropological inspec- tion, have the power to connect Elrst us and later our audience to a group of strangers, a place, a time, an un- folding of events that otherwise seem either too ordi- nary to be worth attention or too extraordinary to be de- scribed at all. Key expressions make the experience of ubeing there" not only real but almost scientiElc (Geertz 1988). Despite my training in linguistics and discourse analysis, I too was attracted to the possibility of repre- senting what were obviously complex social interac- tions through the use of such expressions-although I chose a phrase describing an action rather than a noun describing a concept.

As a fourth-generation linguistic anthropologist, I have had a peculiar relationship with the key expres- sions I used in the field. For one thing, I was not the one who made them up or discovered" them. The Elrst time I went to Western Samoa, in 197S79, I worked as part of a research team. We could count on a number of already existing ethnographies from which to extract key words to enlighten our experience. Our professional heritage was partly shared by our subjects. Many adult Samoans in 1978 knew about Margaret Mead, although only a few had ever read her Coming of Age in Samoa. Our en- trance in the village where we ended up living was facili- tated through a chain of contacts that started with a conversation with fellow cultural anthropologist Bradd Shore a few months earlier. It was from Shore's unpub- lished 1977 dissertation, or perhaps his stories and rec- ommendations at the 1978 annual meeting of the Asso- ciation for Social Anthropology in Oceania, that I took one of his favorite key expressions: teu le va (take care of the relationship). Here is his revised published ver- sion:

Not only are there in Samoan no terms corresponding to the English personality," self" or acharacter," but there is also an absence of the corresponding assumptions about the relations of person to social action. A clue to the Sa- moan notion of person is found in the popular Samoan saying teu le va (take care of the relationship). Contrasted with the Greek dicta Know thyselfn or gTo thine own self be true," this saying suggests something of the difference between Occidental and Samoan orientation. [Shore 1982:136]

After a few months of fieldwork, I started to focus on the discourse produced by chiefs and orators during the meetings of the village council (fono). It is a dis-

course full of fancy metaphors and respectful terms that try to cover or transform (depending on the point of view) the tension that all participants can feel through their bones and hear through their ears. After examin- ing the transcripts of several meetings with long and heated debates about hurt relationships and attempts to mend them, I decided that teu le va was quite appropri- ate for what the matai (titleholders) are up to in the fono. (The expression actually found in the fono speeches was teuteu le va, which means something like make the relationship beautiful," used as an encour- agement to mend the relationship with a nearby vil- lage.) The fono enterprise as a whole could be ade- quately characterized as an attempt to take care (teu) of the relationships (va) among the participants and not only between the village and other political entities. As an event, the fono is a classic conflict-negotiation set- ting, a crucial step in an unfolding "social drama" (Turner 1974) where talk is used to adisentangle" (Wat- son-Gegeo and White 1990) the interpersonal dishar- mony created by past or forthcoming events. Relation- ships must be restored to their ideal if not original condition, to what participants refer to as fealofanz (mutual love, harmony).

Teu le va also became an expression that Elinor Ochs, Martha Platt, and I used in the field to explain our own or other people's behavior. We used it when we tried to convince ourselves that we should or should not do something, trying to come to terms with the tension between our latent beliefs and ideologies and the ethics of a profession that values empathy (or at least neutral- ity) over confrontation. In other words, we found that we were also engaged in taking care of the relationship, the one between us and our hosts, the people we had gone to study.

Many years later, when I started to pay more atten- tion to the discourse of household interactions, among women and children in particular, I realized that, had I decided to concentrate on those settings, I might have chosen a different key expression.

Directives in the Discourse of Caregivers

I remember that during our first field trip I had been shocked by how many orders in the grammatical form of imperatives do this," UgO there," abring that," and so on could be found in the transcripts of the house- hold interactions recorded by Elinor and Martha. As I recount in my 1994 book, this realization was frustrat- ing for a grammarian and discourse analyst looking for acanonical transitive clauses," that is, utterances with fully expressed agents, such as the boy dropped the cup," the woman fed the baby," or the man built a boat." The women's style was also in contrast with the

344 AMERICAN ANTHROPOLOGIST * VOL. 99, NO. 2 * JUNE 1997

men's interaction that I had been recording, especially the oratorical styles of the fono and the ceremonies at- tended by titleholders.

Fifteen years later, as I read through newly col- lected transcripts of interaction between Samoan chil- dren and their caregivers, I Elnd myself dealing with similar situations and a similar discourse, full of orders and elliptical clauses. This time, however, I have not been an analyst of data collected by someone else.

In 1993, Elinor Ochs and I started a three-year proj- ect in a Samoan community in southern CalifoInia, fo- cusing on children's activities, collective problem solv- ing? and literacy tasks (Duranti and Ochs 1996, in press). This time I found Samoans only 30 miles from our house, across several socioeconomic boundaries and close to a freeway exit that I had not explored before this project started. The Samoan families I visited live at walking distance from one another and, more impor- tantly, at walking distance from the Samoan church they all attend (Duranti et al. 1995).

While examining the transcripts of interactions that I had videotaped during one of the visits to a family in our study, I discovered that many of the imperatives in the mother's directives were about locations in space. It was then that I was reminded of the frequent commands in the adult-child interaction recorded in Western Samoa. I realized that, had I started from the in- teractions among children, their older siblings, and adult relatives, I might have been tempted to pick some- thing quite different from teu te va as my key expression, perhaps something more ordinary, less abstract, and yet equally enlightening.

By looking through the 18,000 pages of handwtitten transcripts collected in Western Samoa, I could have seen that one of the most common instructions uttered by adults to children, especially but not exclusively when adult visitors are around, is nofo i lalo!. 1 This ex- pression is composed of a verb (nofo) that describes the body posture of sitting, a preposition (i) roughly trans- latable in English as to, in the direction of," and talo (down, bottom). The three words together are best translated in English with the phrase sit down." Here is an example of its actual use from an interaction where a mother is imploring her oldest son, Niulala, who is 31/2, to act more maturely:

Niulala! Niulala! Niulala! Niulala! Niulala! Niulala! 'oe si kama makua? you Aff boy old (are) you a dear old boy? ngofo lelei X lalo 's? sit good Dir down Tag sit down properly, okay?

Were I feeling the need for a semantic continuity between the fono and the household situations, I could have used an expression I recently found in a dinner conversation recorded in Western Samoa in 1988, when a mother said to her son during a dinner, UTeu lau nofo" (Take care of your sitting).3 Can a relationship (va) be contained, represented, enacted in the act of sitting, in a particular mode of coexistence between one's body and an inhabited surface? In the rest of this article, I argue that it can.

To Hold One's Body, to Know One's Place

Control over the relationship with a lived space seems the paramount preoccupation of adult Samoans trying to control their children's unruly or at least po- tentially inappropriate behavior. Of course, once we look more closely at the situations in which such in- structions are uttered, we realize that ultimately they are about relationships with people and not just with space. For instance, these commands are concerned with the relationship between a child's body's position and the positions of others, especially adults, around him. But adults do not need to be around for the child to be expected to assume a particular posture and position in space. The child is instructed on how to hold his body vis-a-vis other real or potential viewers or bystanders along a vertical axis and a horizontal one. (See Firth 1970 for similar observations about Tikopia postures of respect.) The child's body should be lower or no higher than any other older person's body (vertical axis) and with his legs crossed rather than stretched out (horizon- tal axis). This posture seems to occupy the least room and yet maintains an upward position of the upper body, which must face outward. Figures 1 and 2 illustrate this posture, called fatai or fa'atat, as held by a group f'aukengi) of young caretakers who are attending young babies and making a broom with the midribs of pandanus leaves.

Children who are not supervised and are engaged in relatively stationary activities might take on different positions, such as squatting, as shown in Figure 3, where a group of young boys are playing a fantasy game with make-believe trucks and cement.

It should be no surprise that children's behavior is controlled in terms of the relationship between their bodies and the space they inhabit, and that they are ex- pected to show grespect" Cfa'aaloalo) in this fashion, es- pecially when around adults. Anthropologists have known for a long time that space is often a metaphor for society and that knowing one's place might mean know- ing where one stands in society. This is true in English as well as in Samoan where even high-status individu- als might be reminded of their proper aplace." The term [Transcript, gNiulala," April 4, 1979, book 59]2

INDEXICAL SPEECH ACROSS SAMOAN COMMUNITIES / ALESSANDRO DURANTI 345

Young caregivers sitting cross-legged (fdsataiJ on mats' Western Samoa. Photo by Alessandro Duranti, 1981.

tdlanga means position, place" in both the physical and metaphorical sense. It is used in the fono by speakers who want to scold the behavior of their distinguished colleagues by reminding them of their proper place and, hence, demeanor.

Even Shore's metaphor teu le va, as he points out (1982:311 n.8), contains a spatial reference. The word va, translated as Zrelationship" in teu le va, also means Zspace" or abetween" when followed by the preposition , as in va-t-taimi ("interval," literally Ubetween times") or va-i-niu (interval between coconut trees) (Milner 1966:310). It is easy? then, to hypothesize that the social connotation of Urelationshipt in the word va is derived from, or at least related to, the meaning of a physical space between people or things.

Choosing a key expression that deals with location in space entails entering a rich domain of discourse, one that has been subjected to several lines of analysis (Lawrence and Low 1990; Lynch 1993). In particular distinctions made within linguistic anthropology help clarify how control over who is sitting where and in what posture relies upon and evokes a complex socio- historical world. This world may not be shared by the participants in the interaction, especially when they are

people of different ages and therefore with different ties to and experiences of the Uhome" country-

The Study of Deixis

Traditionally, linguists have been interested in space as part of a larger set of phenomena called deixis. This is the name given to

those aspects of language whose interpretation is relative to the occasion of utterance; to the time of utterance, and to times before and after the time of utterance; to the location of the speaker at the time of utterance; and to the identity of the speaker and the intended audience. [Fill- more 1966:220]

Personal pronouns (I, you), demonstratives (this, that), temporal adverbs and prepositions (now, theKn, before), and spatial markers and expressions (here, there, up, down) are classic examples of deictics (Lyons 1977). Since their meaning shifts from one con- text to the next, these expressions have also been called shifters by Jespersen (1922; see also Jakobson 1971 [1957]).

A Samoan term such as i lalo (at the bottomX down) can only be interpreted relative to a particular person's

346 AMERICAN ANTHROPOLOGIST * VOL. 99, NO. 2 * JUNE 1997

Figure 2 Same group from another point of view. Photo by Alessandro Duranti, 1981.

or object's position in a situation. The request to sit down needs reference to a number of coordinates, in- cluding the addressee's starting position or 07ngo (see Hanks 1990), at the time of the utterance, a position the addressee is asked to abandon, and a target position the addressee is asked to assume.

Deixis, together with the more general phenome- non called indexicality, has received a considerable amount of attention in recent years among linguistic an- thropologists, especially thanks to the work of Michael Silverstein and some of his students. Starting with his 1976 paper Shifters, Linguistic Categories, and Cultural Description," Silverstein extended Charles Peirce's work on types of signs and Roman Jakobson's work on deixis into a theotg of indexicality that goes beyond spa- tiotemporal coordinates to include social reference in both itspresxpposing and creating force. For example, in the English expression this table, the referent of the token of table must be identifiable, must exist cogni- tively, for the deictic this to be interpretable. This is an example of indexical presupposition. In the case of sec-

ond-person pronouns such as you in English, Silver- stein argues that language is the main or even the only medium through which the social category of ad- dressee/recipient is made to exist. Languages that have socially differentiated second-person pronouns, such as the classic T/V type of distinction of many European languages (French tu/vous, Spanish tu/Usted, German du/Sie, and Italian tu/Voi or tu/Lez) are more extreme examples of systems in which words (in this case pro- nouns) are used to activate or establish the relevant so- cial coordinates of equality/inequality, solidarity/power (Brown and Gilman 1960). These are indexes that Sil- verstein sees as maximally creative or performative." As we shall see shortly, the creative function of indexes can also be at work in spatial deixis.

William Hanks's (1990) detailed study of deixis in Maya develops some of these insights by integrating a number of theoretical approaches, including Merleau- Ponty's phenomenological appreciation of the role of the coIporeal field in perception and Irving Goffman's work on frames and participation frameworks. Hanks

I NDEXICAL SPEECH ACROSS SAMOAN COMMUNITIES / ALESSANDRO DURANTI 347

Young boys squatting while playing a fantasy game on the beach, Western Samoa. Photo by Alessandro Duranti, 1981.

underscores that uto engage in referential practice is to locate oneself in the world, to occupy a position, how- ever fleetingly, in one or more sociocultural fieldsX (1990:514). In order to understand spatial expressions, we make hypotheses about the perceptual and cultural horizon against which spatial expressions are used. This honzon, Hanks reminds us, is constituted by the human body, the socially lived space, and the dynamic relationship between individual space and community spaces.

Within a different and yet related tradition of re- search, the work of Charles and Marjorie Goodwin on language, bodies, and material tools highlights different aspects of the lived space, focusing on the ways in which it becomes a resource used by participants in an interaction to test their understanding of each other's actions, words included (see C. Goodwin 1994; Good- win and Goodwin 1996).

These contributions suggest that any expression that makes reference to movement in space must be un- derstood againstperceptually available surfaces, whether

material or conceptual. Thus sit downX has a different meaning in English depending on which target hard sur- faces are available. In a room with chairs, the action of sitting down would be usually understood not only as meaning sit down on a chair" but also as implying some horizontal ad,justment to make sure that the relevant surface of one's body would match the surface of a Useat" or any other Useat-able" artifact. In some cases the question of which chair one might or should sit on might be obvious or indifferent, but in other cases it might be highly contested. It is not unusual to Emd one- self in a situation in which one does not know where to sit and ends up sitting in the wrong" place. I did this re- cently in a doctor?s ofElce, where I responded to Uhave a seat" by sitting in what turned out to be the doctor's chair. The unavailability of artifacts or mediums that (in Gibson's [1986] ecological terms) Uafford" sitting might be interpreted as implying the floor as the target. The criteria by which availability is evaluated are, to various degrees and in various combinations, subjective and context- and culture-speciElc. For instance, is a chair oc-

348 AMERICAN ANTHROPOLOGIST * VOL. 99S NO. 2 * JUNE 1997

cupied by another or by objects a possible target? When a floor and a chair are both available, should one always select the chair? It depends.

In the case of Samoan children growing up in a vil- lage, the instruction to nofo i lalo, for instance, is under- stood differently according to whether they are at school or at home. In a classroom, to sit down" typi- cally means to sit down on one's seat at one's desk.

At home, to sit domrn" means instead to sit on the floor or on a mat that is covering the floor (see Figures 1, 2, and 4). This is consistent with a general expectation that most interactions inside a house will take place among people who are sitting on the floor, or to be more precise, among people who sit on mats that are lying on the floor. Such expectations cover formal and informal gatherings. Figure 4, from a photograph taken in 1978, shows a family having a morning meal. Everyone, even the young boy on the left, is sitting cross-legged on a mat on the floor. Furthermore, as is common in many houses in Western Samoa, there are no chairs or other visible pieces of furniture in this house.

Figure 5 shows a group of chiefs and their wives sit- ting in a Samoan-style house (fale Samoa) during a for- mal Sunday meal (to'ona'i). In this case, sitting down also implies sitting at the periphery of the house floor according to a logic of spatial orientation that pays hom- age to status and rank distinctions (Duranti 1992, 1994; Shore 1982).

Sitting on a mat on the floor is the unmarked posi- tion for a person who is talking with someone else in- side a house. This is conElrmed by a number of cultural practices, including the loud cry 'aumaf fala! (bring in the mats!), which announces that one or more adult visi- tors are entering the family compound. Furthermore, visitors who come inside the house are expected to sit on the floor before introducing the reason for their visit. A visit properly begins after the newcomer has sat down, as demonstrated by the fact that the exchange of ceremonial greetings between the hosts and the guests begins only after the guests have located themselves in a particular spot on the floor, usually a front" section already covered by a seating mat (Duranti 1992).

Samoan family having lunch in a traditional house (fale Samoa), Western Samoa. Photo by Alessandro Duranti, 1978.

INDEXICAL SPEECH ACROSS SAMOAN COMMUNITIES / ALESSANDRO DURANTI 349

Chiefs (right and center) and wives (left) during a formal Sunday lunch (to'ona'i), Western Samoa. Photo by Alessandro Duranti, 1988.

In my experience, the preference for sitting on the floor continues in the nllages in Western Samoa even today, when many houses have chairs or benches. Adults who are under the same roof typically sit down on mats on the floor to engage in protracted conversa- tion, consume food, or engage in certain types of work that mostly need the hands, including weaving and the making of strings or brooms.

Sitting Down in Southern California, or "How Far Down?"

Samoan children living in southern California share some of the activities of their peers on the islands, but they inhabit a very different environment from the one I described above. They too go to school five days a week and to the pastor's school on Sunday and some- times during the week, but they are surrounded by an environment in which the relationship between furni- ture and movement is different from what is experi- enced in an island community. In Los Angeles, houses

are not large open spaces with no furniture or with a few pieces kept at the periphery (see Figure 4), but tpical suburban houses filled with beds, tables, desks, closets, cabinets, chairs, couches, fridges, stoves, ovens, televi- sions, and even video games. Samoan children in south- ern California routinely go out to shopping malls, gro- cery stores, fast-food restaurants, movie theaters, video-game arcades, and toy stores. They also visit friends who are from different ethnic groups, and they are surrounded by a peer culture that is Samoan in their homes and in their church but is otherwise working- class or lower-middle-class American.

When we look through data collected in southern California, we find that the instruction to sit downX is not as pervasive as in Western Samoa. This is in itself an important index of the changing cultural ecology of ev- etyday interaction. I did, however, find that in one of the four families we visited the parents often insisted that their younger children Usit down" and Usit properly." In one videotape (recorded June 7, 1993), over a period of one hour and 20 minutes, I found 14 cases of nofo i lalo and three cases of its polite counterpart, saoya'i i lalo

350 AMERICAN ANTHROPOLOGIST * VOL. 99, NO. 2 * JUNE 1997

shirt lifted while her siblings are busy carrying out other activities and largely ignoring her. Tina attempts a few new moves, doing acrobatics on the floor (14:52) (Fig- ure 8), and then, after her brother has tried again to en- gage her physically by pulling her by the shirt, the mother directs her to sit down. This is phrased as an in- direct request, with the special word saofa'i, a respect- ful verb (RV) usually used in talking about chiefs in their presence. Here it is used, as it is by mothers in Western Samoa, to invoke respectful behavior in a child. This is another case of a creative use of a deictic term, this time a social deictic. 15:18 e le'i iloa saofa'i lelei i lalo?

TA Neg know sit [R] good Dir down don't you know (yet) to sit down properly?

When Tina does not respond, the mother becomes more direct: 15:20 Tina, alu e nofo i lalo.5

go to sit Dir down TinaX go to sit down.

But Tina continues not to comply, taking advantage of the fact that her mother is busy monitoring the activities of the older children, one of whom wants to leave the

Figure 6 Tina dances following rhythm of music on television show (13:32). From video by Alessandro Duranti, June 7, 1993.

(see below). In contrast with other families in the study, asitting down" for this family means that children sit on the floor, even though there is furniture present and the instruction could be reasonably understood by children as meaning to sit on a chair or a couch.

Figures 6 through 9 show a set of body orientations, by a four-year-old girl, Tina, and her Elve-year-old brother, taken from a videotape segment of about Elve minutes, which contains several instructions to Usit down." 4 The sequence starts around 13:25 (min:sec), with disco music being played on television. At this point Tina starts dancing across the floor (see F igure 6). A few seconds later (at 13:53) her brother enters the dance floor with a fake karate move that is not recipro- cated (Figure 7). She continues, lifting up her T-shirt to make it in the shape of a bikini top, presumably imitat- ing the televised dancers (not visible on camera). This attracts the attention of her older sister, seven-year-old Fala, who reports on her to an older male cousin. The mother hears this report and tells Tina to straighten up her shirt cfa'alelei lou ofu.t) (14:04). But Tina does not follow the instruction; she continues to dance with her

Figure 7 Brother playfully "attacks" Tina with a karate move (13:53). From video by Alessandro Duranti, June 7, 1993.

INDEXICAL SPEECH ACROSS SAMOAN COMMUNITIES / ALESSANDRO DURANTI 351

Figure 9 Children finally sit down with their legs crossed (fa'atai) (18:09). From video by Alessandro Duranti, June 7, 1993.

ethnographic description here. There are sufElcient data to make some general observations.

Interpreting the Directive "Sit Down"

There are many questions raised by these data even when presented in a cursory fashion. Let me start by answering the most obvious and perhaps least inter- esting question: Is this all happening because I am pres- ent? This question is partly answered by the fact that {as I discussed earlier) sitting down cross-legged is a pos- ture that is quite widespread in Samoan communities. My presence, which is explicitly referred to a couple of times, seems occasionally to work more like another excuse or reason to get the children to do something they would be asked to do anyhow. It seems much more reasonable to assume that the instruction to sit down is a common strategy in this household, probably used to control the younger children especially, but not exclu- sively7 when other adults are around. Given the limited space in the house alld the high number of children-a total of 11, ranging in age from six months to 18 years it is understandable that the mother would want the physically more active ones to calm down a bit.

Figure 8 Acrobatics on the floor (14:52). From video by Alessandro Duranti, June 7, 1993.

house to go swimming. Two minutes later, their 12-year- old sister, Vae, intererenes. She threatens to tutn off the television or else change channels, and Elnally the mother manages to get some attention.

17:18 nofo i lalo! sit down!

This time the boy sits down, but it will take a lot longer to get Tina to sit down. Although her sister Fala pulls her shirt and tells her aDown!" (in English), Tina sits only when the music ends and the show goes back to dialogue. But even though Tina and her brother have E1- nally sat down, they are not yet in the right" position. The boy is first asked to move back (soso i kua) (17:23) and then instructed to fold his leg properly (noyo fa'alelei la ia mafa'atai 'ou vae).6 Figure 9 shows the fi- nal seating arrangement and postures that satisfy the mother's request.

Although the series of nofo i lalo produced duxing this recording does not end at this point, I will end the

352 AMERICAN ANTHROPOLOGIST * VOL. 99, NO. 2 * JUNE 1997

Is the protracted resistance to follow their mother's request a sign of resistance to maternal authority? Do the children understand the directive when it is said in Samoan? What does it mean to understand it? What is implied by it? The expression i lalo is certainly working as a deictic, as defined above. It is an index that points to a particular relationship between the living body and a perceptually available and yet culturally organized space. As predicted by Hanks's theoretical framework, the indexical expression in this case mediates between different sociocultural spaces. As suggested by Good- win and Goodwin (1996), the interpretation of an in- dexical expression, down" in this case, is a socially ne- gotiated process that relies on existing artifacts, hand gestures, eye gaze, and previous experiences, which are differently known, stored, and reproduced by different participants in the interaction. But what is most striking to me is that the kind of space that is being pointed to is only partly there, and we are left to wonder how in fact such a space could become part of the context" at all. Yes, the children know how to satisfy their mother's re- quest, and eventually, however reluctantly, they do give in and sit on the carpet with their legs crossed (fa'atai). This, they are told, is the proper (or Ugood") way of sit- ting, of showing respect (fa'aaloalo), when a visitor is around and even more so when the visitor is a man with a movie camera who might show the tape to others. But there is more that is being said; there is more that is ex- pected, more to see, hear, and feel.

The mother's instruction Sit down!" has a power and carries a set of implications that go beyond the usual struggle between adults and children coming to- gether at the end of a long day of work, whether in the workplace, at home, or at school. Using Silverstein's classification of indexes, I suggest that nofo z lalo! must be understood as a highly creative indexical expres- sion, one that does more than mediate between genera- tions and conflicting goals that include watching televi- sion, dancing, doing homework, and making decisions. Nofo i lalo! mediates an encounter between an immedi- ately perceivable space and a distant domain of action that is, for the children, invisible and unknown or little known.

In the Los Angeles house, the Samoan mother uses the command nofo (sit) with the indexical expression i lalo (down) to bring order in the middle of the chaotic behavior of restless young children who, after a long day at school, want to watch television, have a snack, dance, roll on the floor, pretend to be karate masters, and engage in other activities that we call being kids." But the spatial domain of such a command is a world made out of houses with no walls, almost no furniture, and floors covered with movable seating mats that have been woven in the same space by women, that is, chil- dren's mothers, aunts, and older sisters. In Los Angeles,

the built environment is different. There is plenty of fur- niture, which is routinely used by both children and adults. And yet the mother's directive must be satisEled by sitting on the carpet on the floor. Despite the fact that the children have not directly experienced the cultural space where the command would have a unique inter- pretation, they know what is meant by it and eventually comply with it.

This is a highly creative use of an indexical expres- sion because it does more than establishing a resting- place goal for the children's bodies. It is an attempt to transfer into the present, and re-create in a very differ- ent context, a distant kind of space. Such a transfer is not only a cognitive one, in which a mental image, held in the mother's memory, needs to be interpreted by the children. It is also the establishment of a social and cul- tural space, which binds the participants by constitut- ing an emotional and a moral commitment to a cultur- ally specific way of being and moving in a house inhabited by other human beings (parents and visitors) who deserve respect. This is the paramount goal of Sa- moan socialization, implying a series of expectations about what it means to be a child and, ultimately, a so- cial being (Ochs 1988).

In conclusion, talking about indexical speech within the framework provided in this article means to pull together a number of implications that constitute the core of a theory of spatial indexicality that must bridge places, identities, and communities. What I have shown here supports the following generalizations:

1. Deictic tokens (shifters, indexes) alone cannot provide the key to their own interpretation.

2. The anchoring of deictic tokens to a speciElc time and place relies on cultural expectations that or- ganize their salience.

3. Different participants might connect linguistic tokens and physical space in ways that show a mutual understanding of the immediate spatiotemporal coordi- nates of the interaction.

4. And yet, participants might have a different reading of what kind of space is being evoked, how far back in time and how far away in physical distance it might be.

What might be simply a here-and-now space for one per- son might simultaneously be a there-and-then space for another a space of memoxy, affect, and duty to invis- ible and yet quite present beings.

The study of such minute and misleadingly innocu- ous details of everyday speech becomes, for linguistic anthropologists, not only a chronicle of how different members of the same family understand the language that is being used in their home but also an index of change. It shows us (and the participants) the extent to which a family is oriented toward the here-and-now as

INDEXICAL SPEECH ACROSS SAMOAN COMMUNITIES / ALESSANDRO DURANTI 353

opposed to the then-and-there. It is no accident that the family where we found the highest number of sitting in- structions is also the family that espouses one of the strongest commitments to thefa'a-Samoa (Samoan way of life) as opposed to the fa'apalangi (Western way). Talking space is a creative act whereby family members can try to infer their identity, while at the same time ne- gotiating their place in history.

Notes

Acknowledgments. Special thanks to James Soli'ai and Elia K. Ta'ase for transcribing some of the segments dis- cussed in this article, and to Benjamin Bailey, Jennifer Reynolds, and three anonymous reviewers for feedback on an earlier draft. Fieldwork on which this article is based was done in collaboration with Elinor Ochs in Western Samoa (in 197S79, 1981, and 1988) and in Los Angeles (in 1993-95). Research in Western Samoa was supported by the National Science Foundation (Linguistics Section) and the work in Los Angeles by the U.S. Department of Education through a grant to the National Center for Research on Cultural Diversity and Second Language Learning, University of California, Santa Cruz. An earlier version of this article was delivered at the 94th Annual Meeting of the American Anthropological Asso- ciation in Washington, DC, November 1995, in the invited session UTheorizing Space: Dialogues across Anthropology," organized by Setha Low; I wish to thank Gary Wray McDonogh for his insightful and encouraging discussion of my paper.

1. In this article, I will use traditional Samoan orthography, including the inverted comma for the glottal stop, [2]. The only two exceptions will be the adoption of the letters ng in place of the letter g for the velar nasal n (I will write talanga instead of tglaga, fa'apalangt instead offazapalagi) and the phonetic symbol n for the transcription of actual speech (see below), which will be quoted between obliques. The choice of ng overg might upset some Samoan purists but should help reduce the mispronunciation of Samoan words by the many readers who are not familiar with Samoan orthography.

The most frequent pronunciation of the phrase nofo i lalo is /nofo i lalo/, due to the fact that in most cases the Samoan spoken in the village is the variety called bad speech" (tau- tala leanga), where the alveolar segments /n/ and /t/, found in the written form (hence in the citation forms used here) and in the speech of most school and church activities, are replaced by their velar counterparts, namely, /n/ and /k/ (Duranti and Ochs 1986). I will also follow Mosel and Hovd- haugen (1992:144) and will not distinguish between the loca- tive i and the directional 'i, which are historically the reflexes of the proto-Polynesian i and ki respectively. This decision is based on a number of factors, including the difficulty that many native speakers currently have making the distinction between the two prepositions and the general tendency not to fully articulate the glottal stop in fast speech.

2. Abbreviations for interlinear glosses: Aff = affective particle, Dir= locative-directional preposition, Neg = nega- tion, [R] = word expressing respect, TA= tense/aspect

marker, and Tag= confirmation particle used in Tag ques- tions.

3. This was actually pronounced /keu lau nofo/. See note 1 for transcription conventions.

4. All names are pseudonyms. The total number of children in this household was 11 at the time of the study. During recording sessions, which took place after the children had come back from school, the three oldest brothers were not usually around. In this family, as is typical of families living in urban and suburban communities, teenage boys were more occupied with after-school sports activities and in general left to be more independent than their female siblings.

5. Pronounced /alu e nofo i lalo/. 6. Pronounced /nofo fa'alelei la ia ma fa'akai 'ou vae/.

References Cited

Brown, Roger, and Albert Gilman 1960 The Pronouns of Power and Solidarity. In Style in

Language. T. A. Sebeok, ed. Pp.253-276. Cambridge, MA: MIT Press.

Duranti, Alessandro 1992 Language and Bodies in Social Space: Samoan Cere-

monial Greetings. Amexican Anthropologist 94:657-691. 1994 From Grammar to Politics: Linguistic Anthropology

in a Western Samoan Village. Berkeley: University of California Press.

Duranti, Alessandro, and Elinor Ochs 1986 Literacy Instruction in a Samoan Village. In Acquisi-

tion of Literacy: Ethnographic Perspectives. B. B. Schief- felin and P. Gilmore, eds. Pp. 213-232. Norwood, NJ: Ablex.

1996 Syncretic Literacy: Multiculturalism in Samoan American Families. National Center for Research on Cul- tural Diversity and Second Language Learning Research Report, 16. Washington, DC: Center for Applied Linguis- tics.

In press Syncretic Literacy in a Samoan American Family. In Discourse, Tools, and Reasoning. L. Resnick, R. Sal jo, and C. Pontecorvo, eds. Berlin: Springer-Verlag.

Duranti, Alessandro, Elinor Ochs, and Elia K. Ta'ase 1995 Change and Tradition in Literacy Instruction in a

Samoan American Community. Educational Founda- tions 94:57-74.

Fillmore, Charles J. 1966 Deictic Categories in the Semantics of Come. Foun-

dations of Language 2:219-227. Firth, Raymond

1970 Postures and Gestures of Respect. In Echanges et communications: Melanges offerts a Claude Levi-Strauss a l' occasion de son 60eme anniversaire . J. Pouillon and P. Maranda, eds. Pp. 188-209. The Hague: Mouton.

Geertz, Clifford 1983 Local Knowledge: Further Essays in Interpretive An-

thropology. New York: Basic Books. 1988 Works and Lives: The Anthropologist as Author.

Stanford, CA: Stanford University Press.

354 AMERICAN ANTHROPOLOGIST * VOL. 99, NO. 2 * JUNE 1 997

Gibson, James J. 1986 The Ecological Approach to Visual Perception.

Hillsdale, NJ: Erlbaum. Goodwin, Charles

1994 Professional Vision. American Anthropologist 96:606-633.

Goodwin, Charles, and Marjorie Harness Goodwin 1996 Formulating Planes: Seeing as a Situated Activity. In

Distributed Cognition in the Workplace. D. Middleton and Y. Engestrom, eds. Pp. 61-95. Amsterdam: Ben-

. .

Jamms.

Hanks, William F. 1990 Referential Practice: Language and Lived Space

among the Maya. Chicago: University of Chicago Press. Jakobson, Roman

1971[1957] Shifters, Verbal Categories, and the Russian Verb. In Selected Writings of Roman Jakobson. Vol. 2, Word and Language. Pp. 130-147. The Hague: Mouton.

Jespersen, Otto 1922 Language: Its Nature, Development, and Origin. Lon-

don: George Allen and Unwin. Lawrence, Denise, and Setha Low

1990 The Built Environment and Spatial Forrn. Annual Review of Anthropology 19:453-505.

Lynch, Michael 1993 Scientific Practice and Ordinary Action: Eth-

nomethodology and Social Science Studies of Science. Cambridge: Cambridge University Press.

Lyons, John 1977 Semantics. Cambridge: Cambridge University Press.

Milner, G. B. 1966 Samoan Dictionary: Samoan-English English-Sa-

moan. London: Oxford University Press.

Mosel, Ulrike, and Even Hovdhaugen 1992 Samoan Reference Grammar. Oslo: Scandinavian

University Press, Institute for Comparative Research in Human Culture (distributed by Oxford University Press).

Ochs, Elinor 1988 Culture and Language Development: Language Ac-

quisition and Language Socialization in a Samoan Village. Cambridge: Cambridge University Press.

Rosaldo, Michelle Z. 1980 Knowledge and Passion: Ilongot Notions of Self and

Social Life. Cambridge: Cambridge University Press. Rosenberg, Daniel V.

1990 Language in the Discourse of the Emotions. In Lan- guage and the Politics of Emotion. C. A. Lutz and L. Abu-Lughod, eds. Pp. 162-185. Cambridge: Cambridge University Press.

Shore, Bradd 1977 A Samoan Theory of Action: Social Control and So-

cial Order in a Polynesian Paradox. Ph.D. dissertation, University of Chicago.

1982 Sala'ilua: A Samoan Mystery. New York: Columbia University Press.

Silverstein, Michael 1976 Shifters, Linguistic Categories, and Cultural Descrip-

tion. In Meaning in Anthropology. K. H. Basso and H. A. Selby, eds. Pp. 11-56. Albuquerque: University of New Mexico Press.

Turner, Victor 1974 Dramas, Fields and Metaphors: Symbolic Action in

Human Society. Ithaca, NY: Cornell University Press. Watson-Gegeo, Karen, and Geoffrey White, eds.

1990 Disentangling: Conflict Discourse in Pacific Socie- ties. Stanford, CA: Stanford University Press.